Yttrium-90 attenuation measurements before and after maximum entropy image restoration
- Univ. of South Florida, Tampa, FL (United States); and others
Quantitative measurement of Yttrium-90, a pure beta emitter, is important for in-vivo management of antibody therapy. Gamma camera imaging of bromsstrahlung radiation poses significant problems, as compared to single photon detection, due to the enhanced scattering and photon penetration effects, and poor conversion efficiency which results in a poor signal-to-noise ratio. For the first time a maximum entropy neural network is proposed for image resolution restoration. It is based on the system MTF, but it voids the common inverse problem associated with both the Wiener and Metz filters. The critical requirement for quantitative measurements is the calculation of an effective attenuation coefficient that is reasonably stable with various source sizes and regions-of-interest (ROI`s) that fully enclose the images of the sources. We have demonstrated that stable image restoration can be obtained for Yttrium-90 with various source sizes and with varying depths in water tanks of 20 cm in depth. Measurements were made using a single-head gamma camera equipped with a high-energy collimator. This yielded effective attenuation coefficients of 0.124 cm{sup -1} and 0.129 cm{sup -1} for the raw images and restored images, respectively. The results for two spherical sources of 3.5 cm I.D. and 6.0 cm I.D. were within {plus_minus}5.07% and {plus_minus}8.84% for the raw and enhanced images, respectively. These results were obtained for ROI`s twice the physical size of each source. In conclusion, the results suggest that sources of varying size, at varying depths in an idealized water phantom, can be imaged with improved resolution using the maximum entropy filter. This filter should permit improved identification of tumors during therapeutic treatment protocols and allow an estimation of the total radiation accumulation in the tumor or in critical organ systems.
- OSTI ID:
- 198026
- Report Number(s):
- CONF-940605-; ISSN 0161-5505; TRN: 95:007029-0167
- Journal Information:
- Journal of Nuclear Medicine, Vol. 35, Issue Suppl.5; Conference: 41. annual meeting of the Society of Nuclear Medicine, Orlando, FL (United States), 5-8 Jun 1994; Other Information: PBD: May 1994
- Country of Publication:
- United States
- Language:
- English
Similar Records
Order statistic-neural network hybrid filters for gamma camera-Bremsstrahlung image restoration
Filtering of SPECT reconstructions made using Bellini's attenuation correction method; A comparison of three pre-reconstruction filters and a post-reconstruction Wiener filter